# Identification and prognostic analysis of propionate metabolism-related genes in head and neck squamous cell carcinoma

## TL;DR

This study identifies four genes related to propionate metabolism that predict survival in head and neck cancer patients and may help guide personalized treatment.

## Contribution

A novel four-gene prognostic model based on propionate metabolism-related genes is developed for head and neck squamous cell carcinoma.

## Key findings

- Four propionate metabolism-related genes (PRKAA2, SLC7A5, GRIP2, CHGB) were identified as a prognostic model for HNSCC survival.
- High-risk patients showed distinct immune infiltration, checkpoint expression, and drug sensitivity profiles.
- The gene signature combined with N stage improved predictive performance for patient outcomes.

## Abstract

Head and neck squamous cell carcinoma (HNSCC) is a highly heterogeneous malignancy with poor overall prognosis. Recent studies have suggested that propionate metabolism-related genes (PMRGs) may play key roles in tumor progression and immune regulation, yet their functions in HNSCC remain unclear.

Transcriptomic data from 502 HNSCC tumor samples and 44 normal tissue samples were obtained from the UCSC Xena database as the training set. Two independent datasets (GSE41613 and GSE6631) from the GEO database were used for validation. Differentially expressed genes (DEGs), key module genes identified via weighted gene co-expression network analysis (WGCNA), and PMRGs were intersected to identify candidate genes. A prognostic model was constructed using Cox regression and LASSO analysis. Immune infiltration, somatic mutations, and drug sensitivity were compared between high- and low-risk groups. Gene expression was further validated by RT-qPCR using clinical samples.

A total of 42 intersecting genes were identified, and four feature genes (PRKAA2, SLC7A5, GRIP2, CHGB) were selected to build the prognostic model. The model effectively stratified patients into high- and low-risk groups with significant survival differences in both the training and validation cohorts. The high-risk group exhibited marked differences in immune cell infiltration, immune checkpoint expression, and cancer immune cycle activity. Mutation burden and drug sensitivity also varied significantly between risk groups. A nomogram combining risk score and pathological N stage showed strong predictive performance.

This study highlights the potential role of PMRGs in immune regulation and tumor progression in HNSCC. The proposed four-gene signature provides a novel tool for prognosis prediction and offers new insights for risk stratification and individualized therapy. Further multicenter validation and mechanistic studies are warranted.